|
|
article
page | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
occur, but the overall ecosystem remains. Can we build networks as self sufficient and steady state as nature?
“Sometimes we technology specialists should leave our desks, forget everything we know about software architecture, computing systems and the complexity of distributed networks and take the time to observe the world around us and learn from Mother Nature. It’s the best self healing system ever made,” Dan Dura, OSS Integration Architect at ‘Cingular now the new AT&T’.
Nature is complex. Way more complex than all the networks together. Yet nature continues to function and even embrace this complexity. If we can understand basic principles of organization and behavior in nature, can we apply these to our simpler networks? Are these principles needed as networks continue to grow more complex?
|
|
|
Most of nature is in constant conflict. It is hard to imagine building a telecom network based on conflict of device against device. |
|
installation of this switch and those supporting devices and systems, to the overlaying of another new service on an existing network structure, and another, and another. Each new element must be linked into the existing network. Seldom do we decommission devices, mostly we keep interconnecting them because they have economic life left or our customers are dependent on them. So maybe there is some useful information we can derive from observing nature and its organization in complexity.
NGOSS is probably the most advanced management and service integration style available today. Yet NGOSS is framework
|
|
|
|
Most of nature is in constant conflict. It is hard to imagine building a telecom network based on conflict of device against device. While nice ordered collaboration in nature is proportionately rare, where you find it, it is very successful: ants, fish schools, plains herds, wolf packs, and mankind. With the possible exception of the higher mammals, what participants of the new science of Complexity are finding, is that seemingly ordered, well designed collaborative behavior can be the results of sets of small rules interacting until they form a complex behavior.
Nature is not designed. All the interactions have been worked out by minute changes in the activities of the individual species acting over a long change. Ecosystems had no team with the job of network designer creating the pretty schematic of the integrated single network. But there are similarities in the way networks grow and ecosystems change. We seldom fork lift an elegant network design into place and fire it up from scratch as a complete working unit. Instead, in reality, networks are ‘grown’ part by part, from
|
|
architecture. Its design was not modeled after nature. “Frameworks capture expertise in the form of reusable algorithms, extensible architectures, and component implementations. Application frameworks have emerged as a powerful technology for developing and reusing middleware and application software. Because current frameworks are application templates, they are not well suited to cope with scenarios with high degrees of heterogeneity, dynamism and unpredictability.” (from Serenity, a project designing autonomic networks) “Patterns capture expertise in the form of reusable architecture design themes and styles, which can be reused even when algorithms, components implementations, or frameworks cannot.” (again from Serenity) Three separate models from nature compete for use as patterns for the design of tomorrow’s networks.
|
article
page | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | |
|
|
